Memory and upload limits not being applied correctly in Nextcloud AIO container

[lionfish2]

Issue Description:
The environment variables NEXTCLOUD_MEMORY_LIMIT, NEXTCLOUD_MAX_TIME, and NEXTCLOUD_UPLOAD_LIMIT are set correctly in the Docker command, but the PHP settings inside the container do not reflect these changes. Specifically:

• memory_limit remains 128M instead of 16384M
• upload_max_filesize is still set to 2M instead of 1024G
• max_input_time is set to -1 (as expected, but no clear reference for expected behavior)

This is affecting the upload performance and overall behavior of Nextcloud, as the limits are being constrained by the default PHP values.

Steps to Reproduce:

1. Remove the old nextcloud-aio-mastercontainer
2. Run the following Docker command to start the Nextcloud AIO container:

docker run -d --name='nextcloud-aio-mastercontainer' \
  --net='bridge' \
  --pids-limit 2048 \
  -e TZ="America/Los_Angeles" \
  -e HOST_OS="Unraid" \
  -e HOST_HOSTNAME="server" \
  -e HOST_CONTAINERNAME="nextcloud-aio-mastercontainer" \
  -e 'APACHE_PORT'='11000' \
  -e 'NEXTCLOUD_DATADIR'='/mnt/zfs-data-pool/nextcloud/' \
  -e 'NEXTCLOUD_MEMORY_LIMIT'='16384M' \
  -e 'NEXTCLOUD_MAX_TIME'='43200' \
  -e 'NEXTCLOUD_UPLOAD_LIMIT'='1024G' \
  -e 'NEXTCLOUD_TRUSTED_PROXIES'='REDACTED' \
  -e 'NEXTCLOUD_OVERWRITEHOST'='example.com' \
  -e 'NEXTCLOUD_OVERWRITEPROTOCOL'='https' \
  -e 'AIO_DISABLE_BACKUP_SECTION'='true' \
  -e 'APACHE_REMOTE_IP_HEADER'='X-Forwarded-For' \
  -e 'APACHE_IP_BINDING'='0.0.0.0' \
  -e 'NEXTCLOUD_MAX_INPUT_TIME'='43200' \
  -e 'SKIP_DOMAIN_VALIDATION'='true' \
  -l net.unraid.docker.managed=dockerman \
  -l net.unraid.docker.webui='https://[IP]:[PORT:8080]' \
  -l net.unraid.docker.icon='https://raw.githubusercontent.com/SpaceinvaderOne/Docker-Templates-Unraid/refs/heads/master/spaceinvaderone/docker_icons/nextcloud_aio.png' \
  -p '7282:8080/tcp' \
  -v '/var/run/docker.sock':'/var/run/docker.sock':'ro' \
  -v 'nextcloud_aio_mastercontainer':'/mnt/docker-aio-config':'rw' \
  --dns 1.1.1.1 --dns 9.9.9.9 --device=/dev/dri:/dev/dri \
  'nextcloud/all-in-one:latest'

3. After the container is running, execute the following command inside the container:

docker exec -it nextcloud-aio-mastercontainer bash
php -i | grep memory_limit
php -i | grep upload_max_filesize
php -i | grep max_input_time

Expected Behavior:
The values should reflect the following:

• memory_limit => 16384M
• upload_max_filesize => 1024G
• max_input_time => 43200 (or as defined by the environment)

Actual Behavior:
The actual values returned are:

• memory_limit => 128M
• upload_max_filesize => 2M
• max_input_time => -1

Logs:
Here are the relevant logs from the Nextcloud server:

{"reqId":"xq93WYgWYWVrLCDEOVZ2","level":3,"time":"2025-06-26T07:59:11+00:00","remoteAddr":"REDACTED","user":"[Username]","message":"Expected filesize of 104857600 bytes but read (from Nextcloud client) and wrote (to Nextcloud storage) 75972608 bytes. Could either be a network problem on the sending side or a problem writing to the storage on the server side."}
{"reqId":"mQugGY7w1SjXbVTAAHKg","level":3,"time":"2025-06-26T07:59:11+00:00","remoteAddr":"REDACTED","user":"[Username]","message":"Expected filesize of 104857600 bytes but read (from Nextcloud client) and wrote (to Nextcloud storage) 80936960 bytes. Could either be a network problem on the sending side or a problem writing to the storage on the server side."}
{"reqId":"cIV2QZrkvZnkNr7lxJbj","level":3,"time":"2025-06-26T07:59:18+00:00","remoteAddr":"REDACTED","user":"[Username]","message":"Expected filesize of 104857600 bytes but read (from Nextcloud client) and wrote (to Nextcloud storage) 79642624 bytes. Could either be a network problem on the sending side or a problem writing to the storage on the server side."}
... (additional log entries omitted for brevity)

Possible Causes:

• There could be issues with the Docker container’s PHP configuration that are not picking up the environment variables correctly.
• The Nextcloud AIO Docker image might be overriding these values with defaults from php.ini.
• There may be other configuration files that prevent these values from being applied at runtime.

[Zoey2936]

they are applied to the nextcloud-aio-nextcloud container, not the mastercontainer: https://github.com/nextcloud/all-in-one/blob/main/Containers/nextcloud/Dockerfile#L127-L133

[cryptic-noodle]

I have the exact same problem with the exact same error in logs specifying that the expected filesize is not the same as the read and wrote filesize.
This happens whenever I try to upload a file larger than approx 150mb. The memory limits in aio dashboard are all default(16G,512mb php limit, 3600s).
I have tried both latest and beta versions and the issue exists in both
My docker compose file is:

services:
  nextcloud-aio-mastercontainer:
    image: ghcr.io/nextcloud-releases/all-in-one:latest
    init: true
    restart: always
    container_name: nextcloud-aio-mastercontainer
    volumes:
      - nextcloud_aio_mastercontainer:/mnt/docker-aio-config
      - /var/run/docker.sock:/var/run/docker.sock:ro
    ports:
      - 8080:8080
    environment:
      - APACHE_PORT=11000
      - APACHE_IP_BINDING=127.0.0.1
      - SKIP_DOMAIN_VALIDATION=true
      - APACHE_ADDITIONAL_NETWORK=proxy
      - NEXTCLOUD_DATADIR=/opt/stacks/nextcloud-aio/ncdata
volumes:
  nextcloud_aio_mastercontainer:
    name: nextcloud_aio_mastercontainer

I have a seperate instance that's running an older version(AIO v10.14.0 and hub 10 v31.0.5) that works perfectly fine.

[cryptic-noodle]

Hey this might help( I use traefik v3 as reverse proxy and turns out it's the culprit)

nextcloud/server#37695 (comment)

[mbehrisch]

I have the same problem. What is the fix for this? I read the answer from @Zoey2936, which is a (great) triage, but not does not lead to a solution IMO.

[mat-l]

Thank you for linking the solution that was helping me.

[divansantana]

@mbehrisch every get a solution? I have the same issue, in the past, with earlier nextcloud-aio versions, zero issue.

Since upgrading (I think to 11.x) I have this issue.

I have a reverse haproxy in front of it. But I didn't make a change to it, and now I have the issue. Only nextcloud updated.
I have respawned the containers with the new memory limits and verified that is fine, but still have the issue.

I have changed the haproxy to have a high timeout values, but still the issue persists.

Not sure how to debug this further. Anyone have any clue?